Voiced speech vs whispered speech: An H2150 positron emission tomography study

By G.M. Schulz, M. Varga, C.L. Ludlow, and A. Braun

Cerebral activation patterns were obtained by PET scans in 20 healthy participants while they told a story with voice (voiced speech condition) or without voice (whispered speech condition) to determine differences in motor control of the larynx, production of voice and the sensory areas that process and monitor the participant’s own voice. The results of this study demonstrated differences in cerebral activation patterns. First, midline areas (ACC, PAG) were only active when the vocal folds are modulated (voiced speech) but not when the vocal folds are held in a tonic abducted position (whispered speech). Secondly, areas in the left temporal lobe were active during voiced speech but not during whispered speech that appear responsible for monitoring one’s own speech production. Third, other regions such as the cerebellar vermis, which may be related to integrating motor and sensory information, were more active during voiced speech than whispered speech. And lastly, bilateral sensorimotor and left primary auditory cortices were more active during whispered speech indicating continuous bilateral cortical sensorimotor activation may be required to hold the vocal folds in a tonic slightly abducted posture to produce whispered speech and the need for increased gain in the auditory system to monitor whispered speech.